[Discussion] Disease

[ptarth]

Reading over the patch notes I found:
If a race -- player or AI -- already has a disease in a city, then their chances of getting another disease go down increasingly with each disease that they have. Hopefully this helps to avoid some of the worst dystopian disease cities, although it won't eliminate them by any stretch. It will make them rarer, anyway.

The situation right now is that you can get a disease it will infect 60% of your population, a turn later a new disease and another 60% of your population, etc. This change would reduce this, but it seems heavy handed.

I would favor a method where the amount of individually infected individuals is scaled based upon the number of wounded citizens you can cure in a round. The more hospital coverage you have, relative to your population, the "earlier" you detect a new disease. This would result in fewer people being infected. In contrast, if you have very little hospital coverage, relative to your population, the larger number of citizen affected.

Example:

The Peltian have 10k population and have 1k worth of wounded citizen repair ability. When they detect a new disease only 500 people have gotten it already.

The Burlusts have 10k population and have 10 worth of wounded citizen repair ability. When they detect a new disease, its because everyone has it already.

The amount might be something like: InfectionRateofDisease * MaximumPopulation/WoundedHealedPerTurn/RacialDiseaseSensitivity

Baseline Disease Sensitivity is some parameter to kludge things into looking nice.

This could be extended to have a disease "hidden" until detected. It would then spread and affect things per normal, but you just don't know why. This would probably be very confusing to players, so while more realistic, it is probably much less fun.

Also on that note: When you have multiple diseases how does it calculate the impairedness of population per building?

What about taking a different tack and having the number of people you can cure scale with vaccine progress and total number of cases seen?

For example, you could have the cure rate (and quarantine/screening effectiveness) start at zero and increase by 1% of your hospital capacity for every ten infections until you get to 100% after a thousand infections. (Perhaps base this on total population rather than static counts.) After that, vaccine research could start kicking in and giving you an extra bonus to cures proportional to the vaccine progress.

That way, diseases could start with smaller initial infected populations and be guaranteed some amount of spread before you get a handle on it. You get a few turns between the disease showing up and the peak impact, which makes it much less of a sudden unexpected disaster, but you still have a major problem that you need to deal with along with everything else going on.

Some further thoughts:

* Doing things the way I suggest gives you a natural scaling of infection with hospital capacity like ptarth suggested. (That's what gave me the idea, actually.) If you have more hospitals, you're better able to deal with the initial cases and head off exponential growth.

* This also interacts in an interesting way with the infection resistance bonus. The lower infectivity keeps more of your population healthy, but it also increases the ramp up time for your cures to become effective. I haven't done the maths, but I think this means that epidemics would have longer but gentler opening phases and then be quicker to mop up when your doctors know what they're doing.

* The decreased cure effectiveness is per-disease, so overlapping epidemics might be in different phases. If disease one is nearing its peak (assuming enough doctors, quarantine, and mortality to cure/cull faster than the infection rate) when disease two hits, the effect is different than if disease one is losing ground or still ramping up.

* This is a similar kind of change to the saucer warp-in mechanic, in that it allows some time for preparation before the full horror is unleashed, but the gameplay effect is quite different.